Background
Left ventricular dyssynchrony and myocardial fibrosis are common findings in patients with nonischemic dilated cardiomyopathy (NDCM). The aim of this study was to investigate the association between myocardial fibrosis and intraventricular systolic dyssynchrony (DYS-sys) in patients with NDCM.
Methods
Thirty-nine patients with NDCM and sinus rhythm were enrolled. Intraventricular DYS-sys was evaluated using Doppler tissue imaging, and cardiac fibrosis was assessed with cardiovascular magnetic resonance imaging with a 17-segment cardiac model. Each segment was graded on a 2-point scale (segmental fibrosis score): 0 = absence of late gadolinium enhancement, and 1 = presence of late gadolinium enhancement. A cardiac fibrosis index was calculated as 17/(17 − sum of fibrotic segments). Receiver operating characteristic analysis was performed to determine the utility of the cardiac fibrosis index to predict intraventricular systolic dyssynchrony.
Results
Patients with DYS-sys had larger left atrial size ( P = .004) and left ventricular end-systolic ( P = .028) and end-diastolic ( P = .034) volumes and lower tricuspid annular Doppler tissue imaging peak systolic velocities ( P = .037) compared with patients without DYS-sys. A cardiac fibrosis index ≥ 1.4 predicted significant DYS-sys with 92% sensitivity and 60% specificity (area under the receiver operating characteristic curve, 0.703; 95% confidence interval, 0.512-0.893; P = .035). Patients with cardiac fibrosis indexes ≥ 1.4 (group 1) had larger left ventricular end-systolic ( P = .044) and end-diastolic ( P = .034) volumes than those with cardiac fibrosis indexes < 1.4 (group 2). Nine of 11 patients (82%) in group 1 and 6 of 28 patients (21%) in group 2 had significant DYS-sys (Pearson’s χ 2 = 12.169, P < .0001). Logistic regression analysis revealed that cardiac fibrosis index ≥ 1.4 (odds ratio, 11.2; 95% confidence interval, 1.72-71.4; P = .012) was an independent predictor of DYS-sys.
Conclusion
Patients with NDCM and prominent cardiac fibrosis have significant DYS-sys. The cardiac fibrosis index is a useful tool to predict DYS-sys.
Heart failure is an important health problem that affects many people. Left ventricular (LV) dyssynchrony, often associated with a wide QRS interval, is common in patients with heart failure. Intraventricular systolic dyssynchrony (DYS-sys) can occur in patients with narrow QRS intervals. Myocardial fibrosis has been demonstrated in patients with nonischemic dilated cardiomyopathy (NDCM) with tissue sampling. Myocardial fibrosis can also be identified as areas of late gadolinium enhancement on cardiovascular magnetic resonance (CMR) studies. Recently, the association of late gadolinium enhancement with adverse prognoses in patients with NDCM was demonstrated. However, there is limited information about the association of myocardial fibrosis and DYS-sys. In this study, we investigated the association between myocardial fibrosis and DYS-sys in patients with NDCM.
Methods
The study population was selected from the patients who were evaluated in the Kartal Kosuyolu Heart, Education, and Research Hospital cardiology outpatient clinic between January 2007 and February 2009. All patients who met the inclusion criteria were asked to participate the study, and those who agreed to participate were enrolled prospectively (39 patients with NDCM and LV ejection fractions < 40%). NDCM was defined by the presence of LV dilatation and systolic dysfunction in the absence of abnormal loading conditions such as hypertension and valve disease, or coronary artery disease sufficient to cause global systolic impairment. The local ethics committee approved this cross-sectional study.
Patients with organic heart valve disease with equal to or more than moderate valvular regurgitation or stenosis, histories of acute coronary syndrome, ischemic electrocardiographic findings, significant (>50% luminal stenosis) coronary artery disease on coronary angiography, permanent pacemakers, and chronic kidney disease (stage > 3) were excluded from the study. All patients were evaluated for functional capacity.
Twelve-lead electrocardiography was performed (0.5-150 Hz, 25 mm/s, 10 mm/mV), and each patient had undergone coronary angiography or cardiac stress testing demonstrating NDCM within the previous 6 months. Standard echocardiographic evaluations with Doppler studies were performed (System 5; GE Vingmed Ultrasound AS, Horten, Norway). LV volume and ejection fraction were measured using the modified biplane Simpson’s method according to the guidelines of the American Society of Echocardiography. The quantification of functional mitral regurgitation was performed using the proximal isovelocity surface area method, as previously described. Effective regurgitant orifice area (in square centimeters) and regurgitant volume (in milliliters) were used as variables expressing the severity of mitral regurgitation.
Doppler tissue imaging was performed in the apical views (4 chamber and long axis) for the long-axis motion of the left ventricle. Two-dimensional echocardiography with Doppler tissue imaging was performed with a 2.5-MHz phased-array transducer. The system was set by bypassing the high-pass filter, while the low-frequency Doppler shifts were input directly into an autocorrelator. Gain settings, filters, and pulse repetition frequency were adjusted to optimize color saturation, and a color Doppler frame scanning rate of 100 to 140 Hz was used. At least 3 consecutive beats were recorded, and the images were digitized and analyzed offline (EchoPAC version 6.3; GE Vingmed Ultrasound AS). Myocardial regional velocity curves were constructed from the digitized images. For detailed assessment of regional myocardial function, the sampling window was placed at the myocardial segment of interest. In the apical 4-chamber view the LV septal and lateral wall, from the apical 2-chamber view the inferior and anterior wall, and from the apical long-axis view the posterior wall and anterior septum were assessed from basal and mid ventricular levels. In this way, septal, anteroseptal, anterior, lateral, inferior, and posterior segments were interrogated at both the basal and mid levels. For the measurement of timing, the beginning of the QRS complex was used as the reference point, and the time to peak myocardial sustained systolic (Ts) velocity was quantified ( Figure 1 ). For the assessment of DYS-sys, the Ts-SD index was used, as described by Yu et al. A Ts-SD value > 33 ms was considered to indicate significant DYS-sys. To assess global cardiac function, the myocardial sustained systolic (s), early diastolic (e), and late diastolic (a) velocities from the basal septal and tricuspid annulus were calculated. Plasma N-terminal–pro-brain natriuretic peptide levels were obtained after 20 minutes of rest following the echocardiographic evaluation, and blood samples were obtained from an antecubital vein. Samples were kept in Vacutainers (Becton, Dickinson & Company, Franklin Lakes, NJ) with ethylenediaminetetraacetic acid and centrifuged for 5 minutes at 1500 rpm. Separated plasma samples were kept at −80°C until analysis. Commercial N-terminal–pro-brain natriuretic peptide assays (Elecsys; Roche Diagnostics GmbH, Mannheim, Germany) were used for plasma N-terminal–pro-brain natriuretic peptide level measurement.
CMR Imaging
CMR studies were obtained with a 1.5-T whole-body scanner system (Magnetom, Avanto; Siemens Medical Solutions, Erlangen, Germany) operating at a maximum gradient strength of 45 mT/m −1 and a slew rate of 200 T/m/s, using 6 anterior channels and 6 posterior channels for data acquisition. Multislice short-axis cine imaging used electrocardiographically triggered, steady-state free precession (slice thickness, 8 mm; interslice slice gap, 20%) acquired from the atrioventricular ring to the apex. The magnetic resonance imaging (MRI) protocol included a functional study of the LV using an electrocardiographically triggered breath-hold segmented steady-state free precession (true fast imaging with steady-state free precession) cine sequence (repetition time, 49 ms; echo time, 1.3 ms; flip angle, 52°) with a slice thickness of 8 mm. For the true fast imaging with steady-state free precession sequence, the temporal resolution was 25 to 39 ms. Depending on the field of view, the typical in-plane resolution was 1.6 × 1.3 mm 2 for all sequences. The total imaging time, including patient positioning, was 45 to 60 minutes. Dynamic studies were performed from the short-axis plane at the basal, mid, and apical segments using 0.1 mmol/kg intravenous gadolinium contrast agent (Dotarem, Guerbet, France) given with a power injector (Stellant; Medrad, Inc, Warrendale, PA). Late enhancement scans were collected in 3 long-axis and all short-axis orientations using a breath-hold electrocardiographically triggered two-dimensional inversion recovery turbo fast low-angle shot sequence (repetition time, 8 ms; echo time, 4 ms; flip angle, 25°). Images were acquired subsequently up to 15 minutes after injection.
CMR images were reanalyzed and documented using the 17-segment cardiac model recommended by the American Heart Association to improve standardization of the results. The left ventricle was evaluated from the short-axis images from basal, mid, and apical segments. The basal and mid cavity were divided into 6 equal segments: anterior, anteroseptal, inferoseptal, inferior, inferolateral, and anterolateral. The apical segment was divided into 4 segments: anterior, septal, inferior, and lateral. The apical cap was termed the apex and constituted the 17th segment. The contrast-enhanced images were analyzed visually by two experienced observers who were blinded to other MRI, echocardiographic, and clinical data. Late gadolinium enhancement was rated by visual assessment, and each segment was graded on a 2-point scale (segmental fibrosis score; 0 = absence of late gadolinium enhancement, 1 = presence of late gadolinium enhancement), using the method of Kaandorp et al because of the frequency of linear and patchy enhancement in patients with NDCM. A cardiac fibrosis index was calculated as 17/(17 − sum of fibrotic segments). Figure 2 demonstrates two patients with and without diffuse cardiac fibrosis.
Statistical Analysis
Statistical analysis was performed using SPSS for Windows version 13.0 (SPSS, Inc, Chicago, IL). Data are presented as mean ± SD and were controlled for normal distribution using the Kolmogorov-Smirnov test and compared using unpaired Student’s t tests when the distribution was normal. A nonparametric test (the Mann-Whitney U test) was used when there was an abnormal distribution. Categorical data between ≥2 groups were compared using Pearson’s χ 2 test. The correlations of continuous variables were analyzed using Pearson’s correlation analysis and categorical variables using Spearman’s correlation analysis. The cardiac fibrosis index was evaluated by receiver operating characteristic analysis to predict DYS-sys. To determine the optimal cardiac fibrosis index value in predicting DYS-sys, the closest value to the best specificity and sensitivity point on the receiver operating characteristic curve was identified. Logistic regression analysis was performed to determine the independent predictors of DYS-sys from the clinical, electrocardiographic, and echocardiographic parameters. A P value < .05 was considered as significant.
Intraobserver (the mean difference between two independent measurements) and interobserver (the mean difference between two independent observers) variability for Ts-SD measurements were analyzed in 10 randomly selected studies and are expressed as the mean percentage error (the difference divided by the number of observations).
Results
The study population included 13 women (33%) and 26 men (66%). The mean age was 45 ± 15 years. Fifteen patients (38%) had significant DYS-sys. The interobserver and intraobserver variability for Ts-SD measurements were 18.2% and 9.8%, respectively. Patients with significant DYS-sys had larger left atrial size ( P = .004), greater LV end-systolic ( P = .028) and end-diastolic ( P = .034) volumes, and lower tricuspid annular Doppler tissue imaging peak systolic velocities ( P = .037) compared with patients without DYS-sys. Patients with DYS-sys had higher Ts-SD ( P < .0001) values than those without DYS-sys. The remaining clinical and echocardiographic parameters were not significantly different between the two groups. The characteristics of the patients with and without significant DYS-sys are listed in Table 1 .
| Significant DYS-sys | Nonsignificant DYS-sys | ||
|---|---|---|---|
| Variable | (n = 15) | (n = 24) | P |
| Women/men | 7/8 | 6/18 | .163 |
| Age (y) | 43 ± 15 | 46 ± 14 | .481 |
| NYHA class | .323 | ||
| I | 2 | 1 | |
| II | 13 | 21 | |
| III | 0 | 2 | |
| QRS duration (ms) | .229 | ||
| <120 | 10 | 20 | |
| ≥120 | 5 | 4 | |
| Log NT-proBNP | 2.71 ± 0.6 | 2.91 ± 0.5 | .275 |
| LA diameter (cm) | 4.6 ± 0.6 | 4.1 ± 0.4 | .004 |
| LVEDV (mL) | 238 ± 71 | 197 ± 44 | .034 |
| LVESV (mL) | 166 ± 55 | 134 ± 30 | .028 |
| LVEF (%) | 30.9 ± 4.6 | 31.7 ± 4.8 | .647 |
| Mitral regurgitant volume (mL) | 19.3 ± 10 | 17.7 ± 8 | .650 |
| ERO (cm 2 ) | 0.12 ± 0.05 | 0.13 ± 0.07 | .464 |
| Mitral E velocity (m/s) | 0.84 ± 0.29 | 0.78 ± 0.25 | .470 |
| Mitral A velocity (m/s) | 0.69 ± 0.24 | 0.64 ± 0.19 | .487 |
| E/A ratio | 1.53 ± 0.9 | 1.35 ± 0.6 | .509 |
| RV DTI s (cm/s) | 9.0 ± 1.6 | 10.4 ± 1.9 | .037 |
| Septal DTI s (cm/s) | 3.6 ± 0.7 | 3.7 ± 0.8 | .728 |
| Septal DTI e (cm/s) | 3.5 ± 1.2 | 4.1 ± 1.7 | .194 |
| Septal DTI a (cm/s) | 4.8 ± 1.6 | 4.8 ± 1.7 | .941 |
| Ts-SD | 46.5 ± 11 | 16.1 ± 9 | <.0001 |
Patients were also evaluated according to the presence of late gadolinium enhancement on cardiac MRI. Nineteen patients (49%) had late gadolinium enhancement on CMR. The most frequently involved cardiac segments were the mid lateral and mid inferior segments, in 13 of 39 patients (33%); the basal lateral and basal inferior segments, in 10 of 39 patients (25%); and the mid anterior and mid posterior segments, in 6 of 39 patients (15%). Linear subendocardial involvement was the most frequent type (8 of 39 [20%]), followed by patchy foci transmural (6 of 39 [15%]), midwall stria (4 of 39 [10%]), and linear subepicardial (1 of 39 [2.5%]) involvement. Patients with late gadolinium enhancement on CMR had higher mitral E-wave velocities ( P = .049) and septal Doppler tissue imaging early diastolic velocities ( P = .015) than those without cardiac fibrosis. Characteristics of patients with and without cardiac fibrosis are listed in Table 2 .
| Fibrosis | No fibrosis | ||
|---|---|---|---|
| Variable | (n = 19) | (n = 20) | P |
| Women/men | 8/11 | 5/15 | .257 |
| Age (y) | 42 ± 15 | 47 ± 14 | .219 |
| NYHA class | .808 | ||
| I | 2 | 1 | |
| II | 16 | 18 | |
| III | 1 | 1 | |
| Intraventricular dyssynchrony | .265 | ||
| Significant | 9 | 6 | |
| Nonsignificant | 10 | 14 | |
| QRS duration (ms) | .640 | ||
| <120 | 14 | 16 | |
| ≥120 | 5 | 4 | |
| Log NT-proBNP | 2.80 ± 0.49 | 2.86 ± 0.56 | .763 |
| LA diameter (cm) | 4.5 ± 0.7 | 4.2 ± 0.3 | .130 |
| LVEDV (mL) | 224 ± 61 | 202 ± 55 | .226 |
| LVESV (mL) | 154 ± 49 | 139 ± 38 | .304 |
| LVEF (%) | 31.8 ± 5.5 | 31 ± 3.9 | .809 |
| Mitral regurgitant volume (mL) | 19.9 ± 10 | 16.9 ± 8 | .349 |
| ERO (cm 2 ) | 0.13 ± 0.05 | 0.12 ± 0.07 | .695 |
| Mitral E velocity (m/s) | 0.89 ± 0.25 | 0.71 ± 0.26 | .049 |
| Mitral A velocity (m/s) | 0.63 ± 0.19 | 0.68 ± 0.23 | .445 |
| E/A ratio | 1.59 ± 0.6 | 1.27 ± 0.9 | .216 |
| RV DTI s (cm/s) | 9.6 ± 2.2 | 10.1 ± 1.5 | .376 |
| Septal DTI s (cm/s) | 3.5 ± 0.8 | 3.9 ± 0.7 | .083 |
| Septal DTI e (cm/s) | 3.3 ± 1.1 | 4.5 ± 1.7 | .015 |
| Septal DTI a (cm/s) | 4.5 ± 1.8 | 5.1 ± 1.5 | .275 |
| Ts-SD | 31.4 ± 19 | 24.4 ± 16 | .235 |
LV walls with fibrotic involvement were also the maximal dyssynchronic segments in 13 of 19 patients (68%). When the 3 most dyssynchronic LV segments were taken into account, cardiac MRI findings revealed that ≥1 of these segments had late gadolinium enhancement in virtually all patients. When we compared the myocardial segments with late gadolinium enhancement with their opposing segments (opposing wall), all segments with late gadolinium enhancement were somewhat delayed in contraction. However, only 12 of 19 patients (63%) met the criteria for significant dyssynchrony.
Receiver operating characteristic analysis was performed to assess the utility of the cardiac fibrosis index to predict significant DYS-sys. A cardiac fibrosis index ≥ 1.4 predicted significant DYS-sys with 92% sensitivity and 60% specificity (area under the curve, 0.703; 95% confidence interval, 0.512-0.893; P = .035; Figure 3 ). The study group was divided into two subgroups according to the cutoff value of 1.4 for the cardiac fibrosis index. Patients with cardiac fibrosis in ≥5 cardiac segments (cardiac fibrosis index ≥ 1.4) constituted group 1 (n = 11), and those with cardiac fibrosis in <5 segments (cardiac fibrosis index < 1.4) constituted group 2 (n = 28). Patients in group 1 had larger left atrial size ( P = .01), greater LV end-systolic ( P = .044) and end-diastolic ( P = .034) volumes, and lower septal Doppler tissue imaging peak systolic ( P = .044), early diastolic ( P = .046), and tricuspid annular peak systolic velocities ( P = .02) than patients in group 2. Nine of 11 patients (82%) in group 1 and 6 of 28 patients (21%) in group 2 were found to have DYS-sys (Pearson’s χ 2 = 12.169, P < .0001). Ts-SD values were also higher in group 1 patients ( P = .004). Five of 11 patients (45%) in group 1 but 4 of 28 patients (14%) in group 2 had wide QRS complexes (>120 ms; Pearson’s χ 2 = 4.322, P = .038). Characteristics of the patients in groups 1 and 2 are listed in Table 3 .
